How do you turn a decision into instructions your AI can build from, and check it built the right thing?
- Module
- 02/05
- To read
- ~8 min
A team asked an agent to “add rate limiting to the API,” and it reported that all tests passed. What do you think they found in production?
In the module you make your guess first, then read what happened and why.
A spec states decisions, not wishes
Acceptance checks are observable and binary
Failing tests first, then code
Plan before code, verify after
A build-ready spec for one feature
A spec for your own app, as a text file you download. The module also gives you the prompt for Claude Code, Codex or Antigravity, a checklist to check what it built, and recommended skills.
- API contracts
- Testing through real entry points